Servo Magazine 12 2006

Tạp chí Servo

Per fe ct ho liday gift s for k id s of all ages !

Per fe ct ho liday gift s for k id s of all ages !

Gift Givers, Take Note

Engineers, We’ve Got

It All!

Enthusiasts, Start Dreaming

Gift Givers, Take Note

Engineers, We’ve Got

Robotic kits help you and your child to experience and learn about perception and

control using a variety of sensors and actuators Challenge yourself with Jameco’sselection of fun and interactive kits! You or your child can assemble the kits andthen enjoy endless hours of discovery

. Robot Insects & Animals . Programmable Robots

. Solar Robots . Educational Kits

. Listening, Touching & Seeing Robots . Legged and Wheeled Platforms

. Hackable Robots . OctoBot Survivor Kit

At Jameco’s RobotStore you can get the world’s most complete robotic offering—

all in one place! You’ll find kits for all ages and experience levels, along with gear

boxes, servos and chassis, for those who are ready to take on the extreme

SERVO Magazine (ISSN 1546-0592/CDN Pub Agree#40702530) is published monthly for $24.95 per year by T & L Publications, Inc.,

430 Princeland Court, Corona, CA 92879 PERIODICALS POSTAGE PAID AT CORONA, CA AND AT ADDITIONAL ENTRY MAILING

OFFICES POSTMASTER: Send address changes to SERVO Magazine, P.O Box 15277, North Hollywood, CA 91615 or

08 Robytes by Jeff Eckert

Stimulating Robot Tidbits

10 GeerHead by David Geer

Robot Dinosaurs Come Alive

and Thrive!

by James Isom with Brian Davis

NXT Robotics: Remote Control

19 Ask Mr Roboto by Pete Miles

Your Problems Solved Here

Baling Wire by Jack Buffington

The Great Serial Port Caper

78 Robotic Trends by Dan Kara

Educational Robotics is the

Smart Choice

84 Appetizer by Robin Hewitt

An Invitation to Computer Vision

with OpenCV

87 Then and Now by Tom Carroll

Bionics — Where Robots Meet

Human Flesh

ENTER WITH CAUTION!

Page 08

Page 10

Page 87

This month: Firefighting robots.

Published Monthly By

T & L Publications, Inc.

430 Princeland Court Corona, CA 92879-1300

(951) 371-8497

FAX (951) 371-3052 Product Order Line 1-800-783-4624

www.servomagazine.com

Subscriptions

Inside US 1-877-525-2539 Outside US 1-818-487-4545

P.O Box 15277 North Hollywood, CA 91615

PUBLISHER

Larry Lemieux

publisher@servomagazine.com ASSOCIATE PUBLISHER/

VP OF SALES/MARKETING

Robin Lemieux

display@servomagazine.com CONTRIBUTING EDITORS

R Steven Rainwater

CIRCULATION DIRECTOR

Tracy Kerley

subscribe@servomagazine.com WEB CONTENT/STORE

Michael Kaudze

sales@servomagazine.com PRODUCTION/GRAPHICS

Shannon Lemieux

ADMINISTRATIVE ASSISTANT

Debbie Stauffacher Copyright 2006 by

T & L Publications, Inc.

All Rights Reserved

All advertising is subject to publisher’s approval.

We are not responsible for mistakes, misprints,

or typographical errors SERVO Magazine

assumes no responsibility for the availability or condition of advertised items or for the honesty

of the advertiser.The publisher makes no claims

for the legality of any item advertised in SERVO.

This is the sole responsibility of the advertiser Advertisers and their agencies agree to indemnify and protect the publisher from any and all claims, action, or expense arising from

advertising placed in SERVO Please send all

editorial correspondence, UPS, overnight mail,

Failure is the path of least

persistence — George M Van

Valkenburg, Jr (1938- )

Robot building is hard work It is

an interdisciplinary craft requiring

expertise in mechanics, electronics,

and programming; each field deep

and wide in and of itself; each has

that “the-more-you-know-the-less-you

know” quality where every answer

creates two new questions and what’s

the point, after all?

Sometimes things turn out better

than expected and inspiration builds

upon inspiration Other times

(mostly), things don’t work out as

planned; both are reasons to have

tried, though (yes, Yoda, there is

“try”) Otherwise, you wouldn’t know

When something doesn’t work or

you don’t have the right part, answer,

or financing, work on another aspect

of the project At least you know what

does not work Enter a contest! This is

a real motivator Nothing like a

deadline to force you to create When

I was building “Autonomous Rodney”

for the 1996 Robot Wars, I couldn’t

get the optical-based passive wheel

disc encoder working and was running

out of time (real robot builders work

best under pressure) Then it

happened in the security section of a

RadioShack Epiphany — magnets

and a reed switch! Worked like a

charm even in the dust at Burning

Man The point is I had dreamed of

building this type of robot for years

but probably would have never

finished it if I didn’t enter that contest

The best laboratory is the real world

and many times we need a deadline to

force us to create The more you get

your robot out and demonstrating it,the better it gets Build, test, anddemonstrate too many folks aretrying to learn everything in the worldbefore actually doing anythingphysical Build, test, and demonstrate

You learn through your fingers

What’s the next step in yourrobotic project? Are you in theplanning stage? Are you “finished?” If

so, show it to someone Be preparedthat some might not be impressedwith your new gizmo, but they aremissing the point of the entire journey

Others will get a kick out of it nomatter how inane Start anotherproject or add to what you have

Robot building is an iterative process

You build upon what you have built

Another problem “The spirit iswilling but the flesh is weak.” Everyonedreams of building a robot but actuallydoing it is too much work Let me saythis robot building is one of themost important things you can do inlife and has more potential payoff thananything Enjoy the pastime and findothers with the same interest

We have had industrial robots fordecades now The new thing ismobility We will soon be entering anage of smart machines where deviceswill know where they are and willeventually pick and place in a clutteredenvironment with great dexterity “Noway” you say? Let’s pretend we’re inthe year 1900 and I was telling youabout the 1950s cars, airplanes,telephones, etc “Not possible!” youwould say Now let’s suppose we’re inthe 1950s and I’m telling you abouttechnology in the year 2000 with theInternet, cell phones, microwaveovens, Global Positioning Systems, etc

Mind / Iron

by Camp Peavy Œ

Dear SERVO:

Regarding a recent Mr Roboto topic I would like tocorrect an apparently widely-held misconception aboutthe transmitted signal of an RC system The desired servoposition is NOT determined by the duration/width of atransmitted pulse The desired servo position isdetermined by the position of a constant width pulse.That is why it is called Pulse Position Modulation Thetransmitted pulse train consists of a start pulse and onepulse for each channel A three-channel system has fourpulses in a frame of data Each pulse is of fixed width,typically 0.25 ms In AM systems, the pulse turns thecarrier wave off Thus, the carrier is on for the majority ofthe time, which helps keep the receiver's AGC (automaticgain control) happy In FM systems, the pulse may increase

or decrease the carrier frequency depending on the brand

of the radio

The position of the pulse — which determines theposition of the servo output — is measured with respect tothe previous pulse The position of the current pulse is thedistance/time between the leading (or trailing) edge ofthe current pulse and the leading (or trailing) edge of theprevious pulse The use of pulse position rather than pulse width minimizes the effect of pulse distortion andlong rise and fall times The transmitted pulses are nottext-book square pulses The rise and fall times areintentionally increased to meet FCC mandated bandwidthlimits With long rise and fall times, the pulse is wider nearthe base than it is near its top Thus, the width issomewhat ambiguous and the measured width maydepend on the signal strength Since the position of apulse is the distance between a point on the current pulseand the corresponding point on the previous pulse, theshape of the pulse has little effect on the measuredposition

The decoder in the receiver separates and converts thisPPM pulse train into individual Pulse Width Modulated(PWM) pulse trains for each servo In some receivers, thedecoder is a serial-in parallel-out shift register Because thetransmitted signal is PPM, there is no delay betweenchannels The pulse for channel 2 starts at the same timethat the channel 1 pulse ends, no gap This can be verified bydisplaying adjacent servo channels on a dual channeloscilloscope

Will Kuhnle Lavon, TX

Writer response:

Thanks for the information This is a bit different from what I have been taught Thanks for pointing these specifics out Perhaps you would be willing to put an article together for SERVO to illustrate these specifics so that this widely-held misconception can be corrected I personally would love to see it — Pete Miles

once again, the inventions seem like science-fiction

Is the pace of technical development slowing? Do you

think there will still be technical obstacles to fully developed

humanoid robots in 50 years? Do you think people won’t

need fully developed humanoid robots? Consider the aging

populations of the developed countries Who’s going to do

the grunt work of the future? Eventually, the robotics industry

will be larger than the computer industry If you go to the

“Computer History Museum” in Mountain View, CA (Silicon

Valley) and follow the evolution of historical computing

equipment, it ends up with robots There will eventually be a

robotics age on par with today’s computer age

The end game for robotics is nothing less than a

humanoid slave Indeed, it is the origin of the word, as

“robot”comes from the Czech “robota” or forced labor

Robots are our progeny They are the next stage in

evolution By 2050, we will have C3PO type androids it

is inevitable If you aren’t building robots today, you are

missing out on all the fun and the other rewards that will

inevitably follow

Mwa-ha-haaa! SV

Did you know that if you’re a paid subscriber

to SERVO Magazine, you can get the online

version for FREE?

Katherine Claire Miles was born October 4th

to proud parents Pete and Kristina.

Weighing in at a healthy 9 pounds, 1 ounce,

she was a meager 20-1/4 inches long.

You done good, Mr & Mrs Roboto!

New UAVs Demonstrated

UAVs for military operations are

becoming ever more common, and Atair

Aerospace (www.atairaerospace.com)

demonstrated two new ones at the

recent Association of the US Army

(AUSA) annual meeting and exposition

First up was the Onyx™ precision guided

parachute system — a parafoil designed

to carry cargo from altitudes up to

35,000 feet, glide autonomously for better than 30 mi, and land on a targetwith accuracy of about 150 feet It combines adaptive control, flocking/

swarming, and active collision avoidancecapabilities to allow multiple systems (50

or more) to work simultaneously in thesame airspace and deliver up to 2,200lbs of “mission critical supplies.”

The nature of such supplies canvary, but a hint is that the Onyx is routinely referred to as a “smartbomb.” Also demonstrated was ascaled-back version of Atair’s LongEndurance Autonomous PoweredParaglider (LEAPP) The Micro LEAPP,which can function autonomously orvia remote control, is designed for special operations intelligence, surveillance, and reconnaissance (ISR)missions that involve up to eight hours

of flight time and a maximum payload

of 50 lbs (Its big brother can spend up

to 55 hours aloft, carry up to 2,400 lbsbeneath its 112-foot wingspan.)

Automatic Refueling

Developed

Also having obvious military cations is the Autonomous AirborneRefueling Demonstration (AARD) system, developed by the DefenseAdvanced Research Projects Agency(DARPA, www.darpa.mil) and

impli-NASA’s Dryden Flight Research Center

(www.nasa.gov/centers/dryden/).

Built on GPS-based relative navigationand an optical tracker, it provides theprecise positioning needed to drop arefueling probe into a 32-in basketthat dangles in the airstream behind atanker (in this case, a Boeing 707-300operated by Omega Air RefuelingServices) and drains into the F/A-18.Although the system initially madethe connection in only two of sixattempts, it safely recovered fromeach flub and completed its mission Inthis demonstration, pilots were onboard the F/A-18 for safety purposes,but the operation was carried outwithout their intervention

With the goal of reducing herbicideuse, Lei Tian, an agricultural engineer at

the University of Illinois (www.

uiuc.edu), has developed a

solar-powered robot that can trackdown weeds and then — using arobotic arm — cut and poisonthem on a close-up and personalbasis The machine — which moves

at about 3 mph — uses GPS fornavigation, plus it sports two smallcameras to give it distance percep-tion An on-board Windows®computer allows it to decide what

is a weed and what is not, and ithas a wireless Internet connectionfor communications and an 80-GBdrive for data storage At present,

The Atair Onyx (top) and LEAPP

(bottom) systems Photos courtesy

of Atair Aerospace.

The AARD system allows autonomous refueling of airborne platforms.

Photo courtesy of NASA, by Jim Ross.

Thanks to the University of Illinois, this solar-powered bot will soon be controlling weeds in some experimen- tal fields Photo courtesy of U of I.

by Jeff Eckert

Are you an avid Internet sur fer

who came across something

cool that we all need to see? Are

you on an interesting R&D group

and want to share what you’re

developing? Then send me an

email! To submit related press

releases and news items, please

visit www.jkeckert.com

— Jeff Eckert

the robot is used only to combat weed

infestation, but in the future, it may

be fitted with different sensors and

cameras that would allow it to examine

soil properties or plant conditions For

now, the device will be used on an

experimental basis, moving along crop

rows in fields at the U of I, but

commer-cial development seems feasible

Motion Card Offers

Reduced Cost

Moving back to the component

level, Performance Motion Devices

(www.pmdcorp.com) has introduced

the Prodigy PC104 Motion Card for

multiaxis, multimotor control Available

in one- through four-axis versions, its

features include trajectory generation,

servo loop closure, quadrature signal

input, motor output signal generation,

performance trace, on-the-fly changes,

commutation, and other functions for

DC brush and brushless DC, step, and

microstepping

The cards are programmed in

C/C++ or Visual Basic and features

include S-curve, trapezoidal, velocity

contouring, electronic gearing, and

user-generated profile modes They

accept input parameters such as

position, velocity, acceleration, and

jerk from the host and generate a

cor-responding trajectory Instantaneous

on-the-fly changes can be sent by the

user, and external signal inputs can beused to program automatic profilechanges Communication is via a PC/104bus, CANBUS, or serial port Prices start

at $380 in production quantities

In a twist that becomes ever moretwisted as you think about it, Hanson

on top of a humanoid robot Hansonhas demonstrated robots that show arange of human expressions, includingjoy, sorrow, and surprise, so you canexpect a fairly impressive level of realism It is suggested that with thistechnology, “You can build yourself tocomfort sick loved ones when you areunable to physically be there or designyourself for posterity.” Sure, and youcould also bring back your belovedAunt Hilda to bake cookies for you

However, it doesn’t take much

imagi-nation to picture Heidi Klum fetching afrosty martini, Barry Manilow cleaningyour bathroom, or Rasputin singing

“Happy Birthday” to you Add a littlemore imagination, and we probablydon’t want to go there SV

R o b y t e s

FaceScan technology will allow robots

to be fitted with heads of yourself, loved ones, or famous people Photo courtesy of Hanson Robotics.

The Prodigy PC104 provides multiaxis

motion control for a range of motors.

Photo courtesy of Performance

Motion Devices.

The technology builder's source for kits, components, supplies, tools, books and education.

Robot Kits For All Skill Levels

Motors, Frame Components and Scratch Builder Supplies ICs, Transistors, Project Kits

BEAM Kits and Components

Books and Educational K

Most orders ship the day received! World-wide shipping Convenient payment options.

Order by Internet, phone, fax or mail.

www.HobbyEngineering.com

1-866-ROBOT-50

1-866-762-6850 1-650-552-9925 1-650-259-9590 (fax) sales@HobbyEngineering.com

180 El Camino Real Millbrae, CA 94030

Visit our store near SFO!

Pleo — UGOBE’s flagship robot —

stands 6.82 inches tall with a width of

5.84 inches and a length of 18.8 inches

Playful, Pleo is capable of responding to

its environment with hundreds of different emotional cues ranging any-where from sadness to grumpiness toplayfulness and everywhere in between

Pleo is the sum of a wide assortment of technology including a32-bit Atmel ARM 7 microprocessor,which is the delightful dino’s primarybrain The prehistoric critter uses a 16-bit sub-processor, which is dedicated to

a camera system’s processing, whichaccomplishes the bot’s eyes’ imageprocessing and bus translation

Pleo also packs four eight-bitprocessors that enable low-level motorcontrol for the robot’s servos, as well

as feedback for “derived” sensors.Speaking of sensors, Pleo is mightysensitive, orchestrating 34 sensors in allincluding the camera

Contact the author at geercom@alltel.net

by David Geer

Robot Dinosaurs Come

Alive and Thrive!

That’s right, it’s robot dinosaurs plural as UGOBE’s Pleo and WowWee Robotics’ Roboreptile make ready for arrival.

UGOBE reserves the right to

change any of the technical details

of Pleo at any time.

Image of Pleo’s remote control dash board.

During its early stages of development, Pleo wasn’t always thehandsome little dino-bot you seebefore you today When UGOBE wasdeveloping its motion systems, Pleowas in a state of … well, you couldsay he was having a bad hair month,but he didn’t even have a head tohave hair on

In that state, Pleo consisted ofpartial body shells and framing with

no skin, some over-used foam andglue, and a piece of duct tape at theend of its neck where its head shouldhave been (the head was beingrepaired) As Pleo’s creators put it,

“Pleo, in this state, was about the ugliest robot you’ve ever seen Butwhen it started to play motions andshow curiosity and emotion throughits movements, it transformed from anugly duckling into a pet — even asstrange looking as it was At thismoment, it really hit UGOBE how

FROM UGLY DINOSAUR-LING

TO BEAUTIFUL ROBOT PET

Specifically, the total

sensor count includes:

• Four foot switches that

detect footsteps up and down

• Seven capacitive touch

sensors for the four legs,

back, shoulder, and head

• A single derived white light

sensor

• Two microphones

• 14-force sensors (one per servo)

• An orientation tilt sensor

• An IR transceiver for bi-directional

data communications

• Another for detecting objects in

Pleo’s mouth (guess they knew kids

would be sticking something in there!)

Additional sensors include those

used to measure battery temperature

and voltage Most sensors are original

designs from UGOBE in order to meet the

specifications for size and compactness

Sensors and hardware empower Pleo’s

intelligence and behaviors Through the

many sensors listed here, Pleo can:

• Recognize objects for avoidance or

• Recognize interactions — like being

touched on its head, shoulders, back,

legs, or feet

• Recognize body position and spatial

orientation and abuse via force

feedback joints (be nice to your new

Pleo, please; he may react if he doesn’t

like how he’s being treated!)

• Recognition of and communicationwith other Pleos (which, of course, youcan only experience if they sell you alot of Pleos)

Pleo also uses 14 motors (standard,low voltage, DC), 150 gears and clutch-

es, a rechargeable NiMH power pack, aUSB port and connector, an SD/MMCslot, and software and systems pro-grammed in a mix of C/C++, Assembly,and open source scripting languages(yet to be announced) Pleo employsUSB communications and a standardfile system for the SD/MMC card

First Ever Robotics and Engineering Appearing in Pleo

In Pleo, UGOBE has combined life-like motion with a wide range

of flexibility of movement, which correspond with and respond in relation to Pleo’s emotional states andwell-being of the moment This isunique in the robotic space

What, When, Where, and How Much?

UGOBE’s Pleo — a new entry in therobot reptile market — will be availablethrough online pre-orders starting inthe midst of the holiday shopping season — depending on your particularholiday — on December 24th InMarch, Pleo will ship to customers whoplace these early orders and becomeavailable in specialty retail stores in limited locations Price tag: $249.Hacking Pleo? UGOBE plans tohave a consumer online SDK and adeveloper SDK — both will be available in 2007 They don’t have allthe details yet for hacking Pleo as

Pleo skinned alive with no skin and no toes.

John Sosoka of UGOBE with Pleo prototype.

Later in development, UGOBE wasbeginning to put on demonstrations ofPleo’s capabilities As they were getting the “motion blending system”

up and running to enable Pleo to perform multiple motions one afteranother, UGOBE was also preparingfor another demo

Working through the night, theywere ready for their demonstration

UGOBE CTO John Sosoka had to catch

a flight the next day to lead the demonstration In the middle of thedemo, where Pleo had just been goingthrough its motions with flawless per-fection, it went into a series of multiplesimultaneous motions, overloading itsjoints and making it appear as if Pleowas having a seizure Sosoka turnedPleo off quickly, as even as a robot, itsmovements were so life-like that itlooked as if Pleo was in genuine pain

PLEO SEIZES UP!

those are still being developed.

Roboreptile

Roboreptile is about 80 cm long by

24 cm wide and 15 cm high in four-leg

mode With batteries installed, it

weighs in at just under 2-1/2 lbs

The Roboreptile from WowWee

Robotics is powered by five motors: two

for the legs and one for the neck, tail,

and jaw By using “high speed resonant

locomotion,” Roboreptile can walk in a

dozen or so different configurations

Roboreptile uses bi-directional

microphones and two IR sensors for its

eyes to detect movement and avoid

objects It has a touch sensor to make

it sensitive to being touched on its

back Through stereo sound sensors, it

can react to sounds in its environment

A light sensor enables Roboreptile

to recognize when its hood has been

placed over its eyes to “calm him

down.” The IR radar also detects,

tracks, and moves toward an IR “food

beacon” to simulate eating

Roboreptile employs a custom-built

RISC CPU with 128 bytes of RAM, a

12K assembly language codespace,and 1/2 meg of sound ROM

Roboreptile Capabilities

All this technology enablesRoboreptile to do some interesting, life-like things In its free-roamingmode, it is hungry and angry (missedits night-time feeding, perhaps?) Inthis state, it explores its environment,avoids obstacles on either four legs ortwo, and makes a lot of angry, hungry-like noises It makes a variety of movements that add to the image ofangry hungry roaming, just like youmight expect to see a dinosaurprogress through in a horror flick

Roboreptile reacts to motionaround it as picked up by its IR visionsensor and responds to sounds by chas-ing whatever is making those sounds

Another method of calming theRoboreptile is to simulate feeding therobot by using the IR remote control

The IR radar detects the feed signaland the robot turns to face the direction that the signal is coming

from The robot even follows the position of the IR signal with its head.Roboreptile then jumps down onits four legs and runs to chase the signal, as if running toward the foodsource When you release the IRremote button, that’s where the robotstops to eat Having been fed, it willmove in a slower, more relaxed fashion,reacting more calmly to sound input

At this point, you can place thehood over its head and it will calm itselffurther Then, it can be picked up andpetted on the back near its touch sensor.Roboreptile — WowWee’s fastestand most agile robot to date — exhibitsfast anaerobic-like motions that areunique to a robotic creation It is thefastest walking robot in its size range.Hacking Roboreptile? For thosewho might want to hack Roboreptile,all the sensors are in the head and thebrain is in its spine All the inputs andoutputs are color-coded and socketedfor easy use and access Motors andgearboxes are double the speed andstrength of earlier WowWee robots.This instills quick reaction times intothe bot It also has two new types ofgearboxes to keep hackers interested.Thanks to UGOBE, makers of Pleo,and WowWee Robotics, makers ofRoboreptile for their fine consumerrobots SV

GEERHEAD

Front view of Roboreptile

and its innards Roboreptile’s remote control in parts Remote control, whole.

Roboreptile with some of his parts bare for the world to see.

UGOBE and Pleo

www.UGOBE.com/pleo/index.html

WowWee Robotics and Roboreptile

RESOURCES

This month, we finish off our Brian Davis series with a look at his remote control The remote uses the Bluetooth capabilities ofthe NXT brick combined with the rotation of a motor to control the power level on the connected robot.

// castling bonuses

B8 castleRates[]={-40,-35,-30,0,5};

//center weighting array to make pieces prefer

//the center of the board during the rating routine

B8 center[]={0,0,1,2,3,3,2,1,0,0};

//directions: orthogonal, diagonal, and left/right

from orthogonal for knight moves

B8

directions[]={-1,1,-10,10,-11,-9,11,9,10,-10,1,-1};

//direction pointers for each piece (only really for

bishop rook and queen

B8 dirFrom[]={0,0,0,4,0,0};

B8 dirTo[]={0,0,0,8,4,8};

//Good moves from the current search are stored in

this array

//so we can recognize them while searching and make

sure they are tested first

with Brian Davis

by James Isom

A bi-monthly column for kids!

LESSONS FROM THE LABORATORY

LESSONS FROM THE LABORATORY

THE CONTROL MODULE

Make this twice.

STEP 2:

Parts:

STEP 3:

downloaded from my website at www.LEGOedwest.com After you have the programs downloaded

to their respective NXTs, simply establish a Bluetooth connection between the two, run the programs,throw the paddles forward, and you’re off and running Have fun! SV

Flip the assembly over.

Over 2,000 Items For Science Gift Giving!

Discover Why the Brightest Minds Think Edmund

Contact Us Today For A Free Catalog

Products That Inspire Science Exploration

For Over 50 Years, Scientifics has been proud to provide you with the latest and best selling, unique science-related items available From unusual gift ideas, to items you just won't find anywhere else–we have something to make you smile… and think.Whether you are a teacher, scientist, hobbyist – or just love exploring the world around you, our unique collec- tion offers a refreshing alternative

Reference CodeCO69-WW12

The IFO 3000 - Magnetic Floating Display

The IFO 3000 takes levitation to new heights From the impressive floating gap to the aesthetically pleasing frame design, the IFO 3000 will grace any home or office The uniqueness of the Floating Ideas Magnetic Display makes it the perfect gift The product is easy to use – the patent pending technology and the spacer training aid that comes with each kit make sure that even a complete novice will be floating straight out of the box The objects rotate serenely in the magnetic field in either direction At the flick

of a switch and a flick and a gentle push of the object, watch as the object slowly rotates forever The IFO 3000 is supplied with an Earth Globe float- ing object as standard Other floating objects are sold separately

Space Shuttle (without base) #P30369-84 $19.95

Nature's Fire™

• Brilliant Moving Sculpture Created by Physicists The dynamics of bringing earth, fire and water together are reflected in this amazing kinetic sculpture Bejeweled lava con- tinuously erupts from the under- water volcano and soft, hypnot-

ic lights enhance the lava's movement in multi-colored pat- terns Lighting is provided by LED, so there are no bulbs to replace It measures 9.7" L x 7.3"

NEW! Cube World

• The coolest new toy of the year!

Discover the distinct personalities and habits of stick figures Slim,

Scoop, Dodger, and Whip These animated stick figures live inside

their own colorful cubes, where they play, think, and sleep Slim uses a

stick to do pull ups and martial arts Dodger bounces a ball Scoop

plays keep away with his pet dog And, Whip jumps rope Watch one

cube separately and play games with an individual figure or connect

the cubes for more interactive fun Surprisingly, stick figures in

connect-ing cubes will wave to each other and transfer themselves between

cubes to fight, play, and even high five Cubes connect in any

horizon-tal or vertical pattern through magnetic sensors on the cube's sides.

Once connected, the possibilities are endless.

Slim/Scoop #P30830-47 $29.95

Whip/Dodger #P30840-23 $29.95

We’ve come to the end of another year of robot

competitions, but the dates for 2007 are beginning to roll

in I expect next year to bring more competitions than

ever as the interest in robots continues to grow Enjoy the

holidays and perhaps you can use the extra free time to start

building robots for next year

Know of any robot competitions I’ve missed? Is your

local school or robot group planning a contest? Send an

email to steve@ncc.com and tell me about it Be sure to

include the date and location of your contest If you have a

website with contest info, send along the URL as well, so we

can tell everyone else about it

For last-minute updates and changes, you can always

find the most recent version of the Robot Competition FAQ

at Robots.net: http://robots.net/rcfaq.html

— R Steven Rainwater

De ce mber

1-2 Texas BEST Competition

Moody Coliseum, SMU, Dallas, TX

In the Texas BEST Competition, students and corporate sponsors build robots from standardizedkits and compete in a challenge that is differenteach year

8-9 South's BEST Competition

Beard-Eaves Memorial Coliseum, Auburn University, Auburn, AL

Regional BEST teams from multiple states compete

in this regional championship

www.southsbest.org

9 ROBOEXOTICA

Museumsquartier, Vienna, Austria

A competition for “cocktail robots” that includesevents such as serving cocktails, mixing cocktails,bartending conversation, and lighting cigarettes

www.roboexotica.org/en/acra.htm

9

Penn State Abington, Abington, PA

Autonomous robots pick up foam balls and shoot

or dunk them in a basket

26-28 Techfest

Indian Institute of Technology, Bombay, India

Micromouse and two other events with the ing names of SNAP and Full Throttle: Afterburn

intrigu-www.techfest.org

Feb ruar y

1-4 Robotix

IIT Khargpur, West Bengal, India

A national-level competition Events includeFastrack Manual, Fastrack Auto, and Softandroid

http://gymkhana.iitkgp.ac.in/robotix

1-4 Pragyan

National Institute of Technology, Trichy, India

Events include TrailBlazer and EyeRobot

www.pragyan.org

26 APEC Micromouse Contest

Anaheim, CA

One of the best-known micromouse competitions

in the United States Expect to see some veryadvanced and fast micromouse robots

Send updates, new listings, corrections, complaints, and suggestions to: steve@ncc.com or FAX 972-404-0269

As a recap, in the October 2006

issue Eric posed a question asking

why so many sensors output a high

signal when it is not detecting

anything, and then goes low when it

detects something.

Q.I believe the reason that many

sensors go low when detecting

something is for safety reasons

Many sensors are used to help in

control-ling and limiting an automated process It

can be really almost anything from

machine tools to mixing chemicals to

detecting particular proteins If the task

being done has limits that must not be

exceeded, then the controller needs to

always know that the sensor is operating

It does this by having the sensor

com-plete a circuit If the circuit is broken by

either the limit being detected or a fault

in the circuit, the controller stops and the

reason for the low state is investigated

Of course, the above method isn’t

perfect, but it is simple and cheap and

is good enough for many uses because

the sensor is never supposed to

actu-ate For example, limit switches at the

extreme travel of a machine tool axis

Just an FYI in case you’re curious

— Eric R Snow

A. Eric, you bring up some

excellent points here Safety and

reliability need to be considered

when wiring sensors into a system The

two things that need to be considered

in determining which way to wire a

sen-sor are: how the system will respond

when it detects a sensor state changeand how the system will respond if thesensor fails Everyone designs their systems considering the first case, buttoo many people forget about the second case They assume the sensorwill always work the way they want it

to Unfortunately, in the real world, sensors fail to work as desired They areeither being used in an application theyweren’t designed for, they malfunction,

or there is a wiring problem betweenthe sensor and the main processor

When working out the wiring logic,one should consider what happens if the sensor fails to work If the output is normally high, how should the systemrespond if there is an electrical failurethat causes the signal to go low? If thesensor’s output is normally low, howshould the system respond if it fails to gohigh when it detects something? As

in your limit switch example, if the controller loses communication to a limitswitch in a machine tool (i.e., it openswhen it is not mechanically triggered),the machine better shut down for safetyreasons But on the flip side, if a sensor’soutput goes low when it detects some-thing, it will go low if the wiring breaks

If a combat robot is designed to attackwhen it detects something, a broken sen-sor would put it into a permanent attackmode which is, in itself, dangerous

Because there are so many ways asensor can be wired into a system,most industrial sensors have both normally open and normally closedcontacts so the same sensor can beused in either configuration

Q. I bought a few of the Sharp

GP2Y0D340K infrared objectdetection sensors because ofyour article in the September ‘06 issue

of SERVO Magazine I managed to get

one apart, but I trashed the adjustablelens I want to use these to detecttrains on a model railroad because Iknow that they will be quite reliable Ibuilt the circuit and it works just fine.Also, the one I destroyed the lens inwill still work because I plan to setthese in the track I can leave the otherones alone because the detector will

be looking up and will work as sold Ihave G gauge so I have plenty of room.Now, my question — how can Imodify the circuit to turn a five-voltrelay on and off when there is a trainpresent? I am a connect-the-dots guywhen it comes to electronics

— John L Deming

A.Sorry to hear that you broke one

of the sensors One would thinkthat if Sharp would advertise thatthese sensors are adjustable, theywouldn’t use such a strong glue A littlepressure is usually all that is needed tobreak the glue, but some parts mayhave a lot more glue than others.Adding a five-volt relay is prettystraightforward Figure 1 shows how thecircuit in the September ‘06 issue wasmodified to control a relay All you need

to do is add a second general-purposeNPN transistor, Q2, and a diode, D2 Thediode is operating as a flyback diode toprotect the transistor from the reversevoltage spike that occurs when the

Tap into the sum of all human knowledge and get your questions answered here! From software algorithms to material selection, Mr Roboto strives to meet you where you are — and what more would you expect from a complex service droid?

relay’s coil is de-energized In this circuit,

I have moved the location of the LED,

D1, to provide a visual indication that the

sensor is detecting something The relay

will be energized when the LED lights

The LED and its current limiting resistor,

R4, are optional They can be removed

from the circuit and it will still operate as

intended The relay can be either a single

pole or a double pole relay, it doesn’t

really matter Hope this circuit will help

with your model train project

Q.I have been an avid reader of

both Nuts & Volts and now

SERVO for quite some time I

have several types of bots and am

currently working on my masterpiece

This one uses a PC/104 main computer

talking to various Atmel controllers

I am trying to find either the

soft-ware or the tools to create a program

that will allow me to send control and

status signals via the Internet from any

remote browser to the robot Once the

control signals are received by the bot,

I would then need to be able to use

these in my control software Status

signals would be sent in the reverse

direction back to the browser I’m currently using Visual Basic as the master control software In addition, Iwould also like to send streaming videofrom the bot back to the browser using

a standard webcam Any ideas?

— Don Peterson Pleasanton, CA

A. This sounds like an exciting and

challenging project The softwarepart of your question is going to

be the harder one to answer, so I willaddress a couple software packages thatmay fit your needs first Since you arecurrently working with Visual Basic, Iwould suggest that you take a look at the

Microsoft Robotics Studio http://msdn.

microsoft.com/robotics/ software, and

the ERSP software from Evolution

Robotics www.evolution.com Both of

these software packages are designedfor PC-based robotic applications, andwork with Microsoft Visual Studio pro-gramming languages such as Visual Basicand C#, and the Net environments

You mentioned that you would like

to use streaming video with your robot

The software from Evolution Robotics

is heavily based on vision recognitionand navigation capabilities They havedeveloped a set of software tools for vision-based applications that work with the Microsoft developmentsoftware, which is quite impressive,especially with their object recognitioncapabilities If vision capability is yourmain goal, then take a serious look atthe Evolution Robotics software

The Microsoft Robotics Studio is anew product under development byMicrosoft, and it is currently availablefor download free of charge TheRobotics Studio is designed for control-ling and communicating with robots,either directly, remotely, or via web-based controls Several roboticscompanies such as Kuka Robotics

(www.kuka.com) and White Box Robotics (www.whiteboxrobotics.

com) are offering the Robotic Studio as

one of the software developmentoptions for their products TheRobotics Studio also has demonstrationapplications for working with LEGOMindstorms RCX and NXT bricks

(ht tp://mindstorms.lego.com/), Parallax BoeBot (www.parallax.com),

470 ohm 2N3904

PIN 5 REG

PIN 4 GND

PIN 6 SHIELD

PIN 6 SHIELD

PIN 1 Vcc

PIN 3 Vout

and the Lynxmotion 6 axis arm (www.lynxmotion.com) Many

of these robots are controlled via serial or wireless controls

With your PC104 computer, you are going to need

software that can run either Windows CE or Windows XP

Embedded (http://msdn.microsoft.com/embedded/) I

don’t know the specifics about your hardware, but the

PC104 hardware developed by WinSystems (www.win

systems.com) has the software for running both Windows

CE and XP Embedded You will have to check with your

hard-ware manufacturer to see if they support these operating

sys-tems and if they are compatible with either the Evolution

Robotics software or the Microsoft Robotics Studio software

Both of these software packages — along with just about any

other software package — will work with sending control data to

the robot and receive status data from the robot This can be

done via a wired serial communication line or via wireless

connec-tion hardware such as the ZigBee wireless modules from

Maxstream (www.maxstream.com) Since you would like

streamless video from your robot using a standard webcam, I

would suggest that you take a look at the wireless webcams from

D-Link (www.d-link.com) and Linksys (www.linksys.com).

Hopefully, the information provided here can get

you pointed in the right direction in developing your

PC-controlled robot When you get your robot up and running,

put an article (or two) together for SERVO on what you did

to get it working Many of our readers would love to know

how you did it, including myself SV

The Tini2131™

The newTini2131™ from

New Micros, Inc.,

comes in the popular Tini

pinout format and is based

on the Philips ARM

LPC2131 The LPC2131

has 32K Flash and 8K

RAM It has on-board

regula-tion, reset circuitry, RS-232

conversion, and three user programmable indicator LEDs

The Tini2131 has 16 of the best I/O pins of the LPC2131

brought out, and separate I2C connections for networking

The 16 shared I/O pins include two 32-bit timers, PWM,

and two serials which can be UARTS, I2C, or SPI There are

up to eight channels of 10-bit A/D This device can be

developed in GCC using Eclipse in the same way as the

original TiniARM, as well as having leading software tool

chains from companies like Keil Software

Its small size — 1” x 1.3” — allows it to be a tightly

integrated solution to robotics, motion, automotive, and

industrial control, as well as the capability of being used in

networking and data logging applications

The Tini2131 features a 60 MHz, LPC-2131 32-bit

ARM CPU The popular ARM processor has wide

third-party language support, free development tools like the

Eclipse development environment using GCC, and a demo

version of the Keil compiler, as well Other languages soon

to come from New Micros are IsoMax/Forth and StatiC for

the ARM As an introductory offer, New Micros will

include a free Keil demo CD upon request

A Tini2131 development kit with the serial cable,

power supply, and proto-development board is available

For further information, please contact:

The IntelliBrain™2

Robotics Controller

RigeSoft has just released the IntelliBrain™

2 robotics controller — its second

genera-the most popular features of genera-the original IntelliBrain robotics controller and the IntelliBrain expansion board on

a single circuit board RidgeSoft has also updated theIntelliBrain-Bot Deluxe educational robot to include the IntelliBrain 2 robotics controller and an ultrasonicrange sensor

The IntelliBrain 2 robotics controller is designed specifically for educational robotics applications Studentsprogram their robots using true Java™ programming —not a Java-like language or other esoteric programminglanguage Tutorials and a course outline — which are available online — facilitate easy integration into computerscience or engineering curriculum

The RoboJDE™ robotics software development environment, which is included with the IntelliBrain 2robotics controller, includes dozens of example programsand tutorials covering everything from basic sensing toprogramming advanced robotic intelligence

The IntelliBrain 2 robotics controller’s design makes

it easy to interface with many popular sensors and effectors including hobby servos, DC motors, infraredsensors, sonar sensors, wheel encoders, vision sensors,compasses, GPS devices, speech synthesizers, and many more

The robotics class library included with RoboJDEprovides an easy-to-use, object-oriented programminginterface to all of the IntelliBrain 2 controller's features, as well as software support for many sensorsand effectors, and an assortment of robotics classes toprovide a foundation for programming intelligentrobots

The IntelliBrain 2 robotics controller includes the following features:

• Java programmable

• Two DC motor ports

• Five servo ports

• Seven analog/digital input ports

• 13 digital input/output ports

• Two RS232 serial ports

New Micros, Inc

The IntelliBrain-Bot Deluxe educational robot includes:

• IntelliBrain 2 robotics controller

• Two servo motors

• Two wheel encoder sensors

• Two line sensors

• Two infrared range sensors

• Ultrasonic range sensor

• Chassis, wheels, and required hardware

The IntelliBrain-Bot Deluxe educational robot kit can

be purchased either assembled or unassembled

For further information, please contact:

Brushless DC Motors

DurA-Tek® brushless

DC motors available

from AMETEK® Technical

& Industrial Products

feature integrated drive

electronics enabling

enhanced motor

con-trollability in a smaller,

lighter package Their

3.0-inch outside motor

diame-ter and standard two-wire electrical hookup further allow

them to replace similarly sized brush-commutated DC

motors while delivering relatively higher performance and

extended service life

These motors ideally suit demanding

high-duty cycle applications for equipment used in the

transportation industry Applications expand into

HVAC, chemical, mining, medical/biotech, data

storage, semiconductor processing, automation, and

other industries that can benefit from compact

and rugged motor construction For harsh

environments, these motors have been designed to

resist hot-water spray, rain, humidity, salt, fog, shock,

and vibration

“Smart” onboard motor controls and advanced

electronic design deliver key features, including

multi-speed operation, over-current control,

locked-rotor protection, reverse polarity protection,

transient over-voltage protection, and over-temperature

shutoff

These 12V/24V brushless DC motors can achieve

continuous torque from 19.7 oz.-in to 55 oz.-in and

speeds from 2,927 RPM to 4,400 RPM, depending on the

model Custom products can be engineered

For further information, please contact:

Control Your R/C Vehicle or Robot From Your PC

Endurance R/Coffers an inter-face system calledthe PCTx Thisdevice allows auser to control aradio control vehi-cle or robot via a

PC The PCTx wasdeveloped with the intention of providing users with a lowcost means to achieve wireless control via a PC Also, byutilizing hobby transmitters, no modifications to the vehi-cle or robot are necessary in order to achieve PC control.The software required to operate the PCTx has been openly released on their website Endurance R/Cfeels this allows for a greater level of innovation and also allows many new applications to be developedaround the system

The PCTx requires a radio transmitter with a buddybox/trainer port and a Windows PC with USB port

A universal version for pistol grip style transmitters is currently in the works PCTx technical specs include:

• Supports up to an eight channel radio system

• Buddy box/trainer port enabled radio required

• Pulses refreshed at 50 Hz

• Independent servo control on all channels

• C++ software API available, VB coming soon

AMETEK Technical &

us to run in our New Products section, please email a short

description (300-500 words) and a photo of your product to:

newproducts@servomagazine.com

Show Us What You’ve Got!

PO Box 482 Pleasanton, CA 94566 Email: info@ridgesoft.com Website: www.ridgesoft.com

RidgeSoft, LLC

Halloween R Robot T Terror 200 6

The first ever Halloween Robot Terror is over and

everyone had tons of fun There were two brand new teams

competing in their first ever event Team Bad Bots with their

bot Black Wedge is from Palo Alto, CA The second new

team is un-named as of yet with their bot Screamer, a

modified BB toy Also Team Misfit has a new builder/driver,

Dan, driving a flea weight named Atom Bomb Welcome to

the sport guys! Hope you have as much fun as the rest of us

The costume contest was a great success and had nine bots

compete I kept hearing a lot of builders saying “next year I’m

going to do ” so I will be doing the Halloween Robot Terror

again next year The winners of the Bot Costume contest are:

• First Place — Stumpy from Team DMV

• Second Place — Front Kick from Team Kick-Me

• Third Place — Scream (the brand new team that’s un-named

for now)

Photos are posted on the CIB website at www.cal

bugs.com.

In the Flea Weights, there were only two competitors, so

I fought them for the best of two out of three:

• First Place — Change of Heart from Team Misfit driven by Kevin

• Second Place — Atom Bomb from Team Misfit driven by Dan

In the Ant Weights, I had seven bots competing:

• First Place — Fire Eagle from Team Misfit driven by Kevin

• Second Place — Stumpy from Team DMV driven by David W

• Third Place — Pooky from Team ICE driven by David L

In the Beetle Weights, I had three robots competing, so

I ran them round robin

• First Place — Toe Poke from Team Kick-Me driven by Hugh

• Second Place — Unknown Avenger from Team ICE driven byDavid L

• Third Place — Itsa from Team Bad Bot driven by Mike

Dave Wiley Bot Gauntlet Baron

Continued on Page 84

Third place costume winner Screamer driven

by Gabriel His team is so new its not even named yet

This is team DMV Ant Weight Stumpy in costume The eyes light up and the head turns left and right Also, the handle bars turn when Stumpy turns left and right This costume took first place

Rosie has gotten a LOT of use at several events by people from the audience and it’s easy to get a line of people wanting a turn to drive her around This helps keep the fighting sur face very clean

Costume Contest Winners

Featured This Month

30 Results — Sep 12 - Oct 13

32 Upcoming — Dec and Jan.

Technical Knowledge

31 Radio Systems

by Leonard G Ginn, Teampyramid

Product Review

32 DuraTrax IntelliPeak AC/DC

Mini Pulse Charger

by Kevin Berry

The subject of arena design isalways good for a discussion,debate, and (usually) a disagree-ment One thing no one disagrees on, however, is that thearena is the first and last line ofsafety between hard, sharp botsand soft, squishy people Despite

a few close calls over the years,the arenas built and used inCombat Robotics have a sterlingrecord of protecting people fromharm

The first rule of arena design

is to assume it must contain the next class

bigger thanyou’re planning

to fight in it So,

if you’re ing for the 30pound “feath-

build-e r w build-e i g h t ”class, yourarena should

be able to

w i t h s t a n dthe weaponloadout of a 60

pound “lightweight.” An insectarena for ants and beetles should

be able to contain a Mantis oreven Hobbyweight weapon This

is above and beyond whateversafety factors are built into thebasic design

Most arenas are built from acombination of polycarbonate(“Lexan” is a commonly usedcommercial term for this), alongwith steel, aluminum, and wood.Insect class arenas need at least 1/4” poly, and big botboxes range from 1/2” to 1”

depending on design and weight

class Insect arenas run from a tight

4’ x 4’ to a roomy 8’ x 8’ in size,

while large bot boxes are 12’ x 12’,

up to 32’ square They tend to

run in 4’ increments due to the

availability of polycarbonate in 4’ x

8’ or 4’ x 12’ sizes Ceiling heights

are generally 4’ for insect bots,

and 8’ in large boxes, for the same

reason

Keeping the bots off the walls

is a key design feature, and most

arenas incorporate either fixed or

“loose” barriers such as “I” beams

or railroad ties to do this Ceilings

are often plywood, sometimes

covered with moving blankets or

other debris absorbing media

Doors can be a weak point, since

they must be opened and

closed twice for each fight, but

they still need to provide the same

level of protection as the rest of

the box

Setup, teardown, and

trans-portation also figure into the

design equation Insect boxes can

(usually) be set up in a couple of

hours, while large arenas might

take two days

Another design factor is nance After almost every event,some sheets of poly need to bereplaced Also, floors take a beating

mainte-Even during events, parts of thearena may need to be repaired

or replaced Other factors includelighting, ventilation (especially ifinternal combustion engine (ICE)bots are allowed), audience visibility,driver visibility, and what type of surface the arena will be used on

Setting up a box on a concrete slab

is a much easier task than on dirtand grass!

This brief overview isn’t intended to take the place of athorough design discussion on current or new arenas However, as

a recurring topic on forums andbulletin boards, it’s obviously animportant one Maybe it’s timeeach organization in the sport took

a look at their arenas, along withthe “arms race” of increasinglypowerful weapons, and ask themselves some pointed safetyquestions, like “is our box still safeenough?” SV

Combots arena showing drivers stations Roof structure is built with trusses Spanning large arenas is sometimes a difficult design feature.

The TC Mechwars arena features a clear roof

to increase visibility in large venues.

Inside the Combots arena, showing bumpers designed to keep bots away from the walls Spinning weapons just a few feet from spectators require lots of thought

into a safe design.

“Toad Tank” showing roof support and steel box beam bumpers.

The new SECR Florida insect box being assembled This arena was built using 80/20 aluminum products, 1/4” polycarbonate, and a plywood floor with sacrificial hardboard overlay.

Inside the Battle Beach box Back walls

are steel; railroad ties keep bots off the

polycarbonate Overhead lights help with

visibility Photo is from perspective of

drivers behind protective window.

The WarBots arena has a classic setup, with drivers platforms and a ramp in between.

Team Whyachi’s arena has small boxes to

allow testing the next match’s bots while

fighting happens in the main box.

Weapon safety locks are one of

the best ways to stay safearound combat robots They provide

the last element of protection

between you and the robot when

precautions fail or mistakes occur

The purpose of a weapon lock is to

prevent the motion of any weapon

with the capability of dangerous

movement

Three Basic Criteria

A weapon lock should meetthree basic criteria It must: prevent

dangerous movement of the

weapon; be clearly visible in shape,

size, and color; and be able to be

removed and inserted quickly and

easily The removal or insertion of the

weapon lock is the most important

criteria aside from effectiveness and

is the last action that should be

performed before leaving the arena

and the first when reentering the

arena Every second that is wasted

while trying to remove or insert the

lock, leaving the weapon to move

freely, is one more second where afatal mistake or error could occur

When it’s Necessary

A weapon lock is necessaryevery time a robot is turned on;

regardless of the situation Why?

Mistakes can and do happen andwhen it’s yours or other people’ssafety on the line, it’s worth the extra effort

For example, when testing thelifting arms on my 30-lber, I checkedthe radio settings and turned therobot on The arms instantly activated as a result of plugging themotor into the wrong receiver port,causing dangerous unexpectedmovement Remember, always useyour safety lock!

Techniques/Methods

While every robot is different,the most effective and easiestweapon lock is a simple pin or bar

Spinners often use a hole that is

designed into the spinning weaponalong with a matching hole in theframe for a pin to drop into Anotheroption is to put two holes on eitherside of the weapon so that insertedpins will prevent motion in bothdirections An excellent example of aweapon lock is shown in Figure 1.For every case, there is a differentbest way to secure your weapon Thetask then is to simply use commonsense and find that best way that satisfies the three basic criteria

What to Avoid

Many builders leave the weaponlock as a last thought, something tofigure out after all the “important”stuff is correct However, this oftenresults in locks that do not functionsafely Improper examples ofweapon locks include attachmentsthat require they be clamped, bolted

on, or screwed in before theybecome effective A perfect example

of what not to do is shown in Figures

2 and 3 SV

When my son Karl was still in

fifth grade, we started buildingcombat robots together It just

seemed like a natural progression for

a kid whose first word was “broken,”

and who got a sledge hammer forhis third birthday

We had a lot of fun with this, sowhen he was in middle school, Iwent to the organizers of the local

after-school program with an offer Isaid, “Hey, I am willing to teach aclass where we let hyperactive sixthgraders build heavily-armed remotecontrol juggernauts.”

Weapon Safety

● by Brian Benson

FIGURE 1 Ziggy of Team CM Robotics uses

a steel bar to secure their system in the

extended and retracted (not shown) position,

preventing any dangerous motion The weapon

can output a force of 14,100 lbs, so they use a

bar capable of handling 80,000 lbs of force.

FIGURE 2 This is a spinning weapon being held only by a clamp — exactly

what should not be done.

FIGURE 3 Even small bots need thought put into weapon restraints This Mantisweight restraint is definitely not safe!

● by Tim Wolter

They thought that was an

absolutely appalling notion, but

after some discussion, we

com-promised on a smaller scale

ven-ture that has been running with

success for the last five years

We take classes of 24 kids

at a time and have them each

build an R/C combat robot of

one or three pound size The

“final exam” for each class is an

all-out tournament where the

students employ their engineering

and driving skills to try and reduce

their opponents to smoking rubble

And we do this with a near zero

budget and a high level of safety To

accomplish this, we “freeze” the

technology level so that it is

affordable, and so that all students

compete with similar equipment

The most basic machine utilizes

two Hitec servos which the kids

“hack” for 360 degree excursion and

couple to drive wheels These can be

hooked to a standard R/C receiver

and a six-volt NiCad or NiMH

battery to make a basic “pusher” or

wedge robot

Most students want to build

something just a bit fancier With a

bit of soldering, a four-wheeled

machine can be built by coupling two

servos per side And active weapons

can be added using 9.6 volt R/C car

drive packs connected to motors

driving spinning bars or discs

Weapon control is via a micro servo

that closes a circuit as a simple

mechanical relay, or with low-cost

airplane electronic speed controllers

I make a point of encouraging

far out designs Other than the

weight classes, the only rules are:

1 No flame throwers

2 No hand grenades

3 No live animals

So, we have had robots built

from old Nintendo controllers,

sponges, wood, and mixing bowls

They have featured armament

ranging from six inch circular saw

blades, to a hammer bot wielding a

stone arrowhead, and more than a

few machines that go into battleequipped with little more than

a good paint job and ill founded confidence

The cost of the program is low;

each student pays a $20 class fee,which we waive for financial need

Hitec servos — usually 311 or 325HDs — run about $8-$10 each

HS-An excellent source for these can be

found at www.servocity.com I also

usually have a few donated machines

or wreckage left over from previousclass sessions for kids who want totry a four-wheel version I have con-nections that provide me with Lexanscrap that comes in handy And forthe electronics, I have been graduallyaccumulating transmitters, receiversand batteries off of eBay, from donations, and from various low cost

sources, such as www.battery

space.com Tom’s RC (www.tti-us.

com) has some good electronics, as

does the Robot Marketplace

(www.robotmarketplace.com) As

the world of combat robotics is heavily populated with, well, techno-nerds, it is not surprising that theInternet is the necessary glue thatlinks the community together

There are also many useful components to be found for free Igenerally have one session I call “junkday” where kids are encouraged tobring power tools, toys, small appli-ances, and such to be cannibalized Ihope that these are non-functionalitems, and that parental permissionwas granted! It has become neces-sary to specify no VCRs; with theadvent of the DVD player, there arepiles of these sitting around, andthey yield few useful components

To do a class of this sort, it isnecessary to break the group up intotwo groups of 12 students each.That is about the maximum number

of adolescent goofballs that can becontained in one place and watchedwhile they use tools Each class getsabout eight sessions of roughly 1.5hours of build time Interestingly,while my classes have been about 95% male, the girls who doparticipate always build outstandingmachines

So far, the middle school hasbeen supportive We get a tech edclassroom to work in, a certainamount of storage space, and achance to do our end-of-class tourna-ment at the school on a Saturdayafternoon Teacher participation hasbeen helpful but intermittent, so theshow is usually run by myself, myson, and various parent volunteers.Since the school shop is usuallynot available, we bring our owntools, which include a soldering iron,cordless drill, small drill press, and asheet metal shear for cutting Lexan.Each kid brings a shoebox to containhis work in progress

Our school has a strict noweapons policy that probably banseverything beyond plastic forks, so Ihave kids bring any dubious itemsdirectly to the class storage area Sofar, nobody has had any problems onthe bus to school

The goal of the class is to havefun, but along the way, we do manage to teach the kids a bit aboutdesign, use of materials, radio frequencies, wiring circuits, and use

of some basic tools With close attention to safety glasses and a ban

Assorted three pound competitors The hammerbot

in the foreground actually wields a stone arrowhead!

Frantic battery charging.

on unsupervised weapon activation,

we have had no injuries worth

mentioning We apply about one

band aid a year, usually to some

part of me

The tournament is a majorundertaking I have a 6’ x 6’ Lexan

enclosed arena with a plywood

floor Arena hazards vary with my

whims, but usually include a saw and

a grinder, along with a trap door

Arena hazards are run off a control

board with audience volunteers,which are never in short supply

We are not yet at the pointwhere we have 24 sets of electronics, so hard-working volun-teers continuously swap receiversand batteries from one machine toanother using Velcro and plenty ofduct tape Local R/C enthusiasts andcombat builders from the Twin CitiesMechwars group have always beenthere to pitch in

We usually weld up trophiesfrom whatever junk is lying aroundthe workshop, and award them forfirst through third in each of theweight classes There is also a specialaward for best design, by which

I mean most innovative, not sarily most successful The highest

neces-award of all is the coveted “GoldenDumpster,” given for most effectiveuse of junk materials We live in atime when hands-on tinkering is lesscommon than it once was, and Ibelieve in encouraging kids to scavenge and experiment as much aspossible

With the generally declining cost

of electronic components, the class is

in some ways getting easier to runover time But looking ahead I suspect that I will at some point give

in to the temptation to have the class start building more advancedmachines, perhaps with antweightcontrollers such as the Scorpion, oreven 12 to 15 pound machines forone of the several competitions thatrun in our area SV

Emergency repairs. Picachu — a surprisingly effective machine

fashioned out of old RC car bodies. A possible world record for percentage composition of duct tape.

Fall W.H.R.E

’06 was held

9/16/2006 in

Dorchester, WI

Results are as follows:

• Heavyweight — 1st: “Ty,” plow,

Bobbing For French Fries; 2nd:

“Brick,” wedge, Whyachi

• Hobbyweight — 1st: “Ricochet,”

wedge, Whyachi; 2nd: “Shroom ofDoom,” wedge, Delta Strike Force2001

• Beetleweight — 1st: “Firefly,”

wedge, Booyah; 2nd: “Jeepy Jeep”;

3rd: “3A,” spinner, Whyachi; 4th:

“Celebrity Lunchbox,” ram, LovBots

• Antweight — 1st: “Underwhere?!,”

spinner, Hazardous Robotics; 2nd:

“ANTI,” spinner, 564 Robotics; 3rd:

“Nano Falcon,” drum, Whyachi; 4th:

“Wykydtron,” spinner, Delta StrikeForce 2001

• Fairyweight — 1st: “Kankle Killer,”

spinner, Whyachi; 2nd: “Destroyer

of Grass.”

Robothon Robot Combat 2006was held 9/30/2006 in Seattle,

WA Results are as follows:

• Hobbyweight — 1st “Shear Terror,”

spinner, Sparkle Motion; 2nd:

“Fiasco,” spinner, Velocity; 3rd:

“Scratch,” clamp, Gausswave

EVENTS

RESULTS — September 12 - October 13

• Beetleweight — 1st: “Hurty Gurtie,”

drum, Death By Monkeys; 2nd:

“Creepy Crawler,” wedge, X-Bots;

3rd: “Altitude,” spinner, Velocity

Marin Ant Wars VI was held

9/30/2006 in Tiburon, CA

Results are as follows:

• Antweight —

1st: “The Bomb,” drum, Misfit; 2nd:

“Emsee Fry Pants,” drum,Burntpopcorn.net; 3rd: “MC PeePants,” drum, Fatcats; 4th:

“Unblinking Eye,” spinner, HammerBros

• Fairyweight — 1st: “Microdrive,”

lifter, Misfit; 2nd: “Hulk Hogan,”clamp, Fatcats; 3rd: “Crisp,”flamethrower, Offbeat; 4th: “Catch22,” drum, Hammer Bros SV

So, you are going to build a

combat robot One of the most

important things to think about is the

type of radio system that you are

going to need If you are building a

large bot class (more than six lbs.),

you will need a PCM (pulse code

modulation) type radio system or

better For the small Insect robot

class (six lbs or less), you may use a

PPM (pulse proportional modulation),

or you may get by with a toy radio

system This may also depend on the

rules of the events you plan to attend

with your bot So check before you

invest The radio system has two

parts: the transmitter, which you hold

in your hands; and the receiver, which

goes in the bot The radio can be a

trigger style or twin stick

Most electronic speed controls for

combat robots only have two or three

channels, with sometimes a fourth

“invert” function The twin sticks

radio’s have two or more channels (up

to 6-8), depending on the type With

the twin stick system, you can drive

the robot with tank turn steering

Typically, radio systems can come in

five different frequencies They are:

27, 49, 50, 72, and 75 MHz

The new radio system on the

block now is the 2.4 GHz This unit,

when first activated, searches 80

channels in the 2.4 GHz band When

the radio locates the best clear

channel, it will lock on to that

channel The 2.4 GHz frequencies

are so much higher than those of

conventional channels that yourradio won’t even recognize the existence of the other bands

The 50 MHz radio is a problembecause it requires you to have anamateur radio license from the FCC

The 72 MHz systems are for aircraftuse only, and are not allowed forcombat robots The toy radio systemsare 27 MHz and 49 MHz With 27MHz, you may have a choice of different channels from A1 to A6,and 49 MHz has only one usablechannel There are also many higherquality 27 MHz systems used for R/Ccars and boats The 75 MHz systemuses channels 61 through 90 You do

not need a different radio to changechannels; you can simply change thecrystal in the transmitter and receiv-

er You will, however, need to checkwith the radio manufacturer on this.The 75 MHz radio system comes in three different bands:

AM (amplitude modulation), FM (frequency modulation) and FMPCM The AM radio range is extreme-

Radio Systems

● by Leonard L Ginn, Teampyramid

Trigger Style — Futaba Magnum Sport FP-T2PB,

two channel.

Tower Hobby two channel AM radio.

Spektum DX6 six-channel DSM 2.4 GHz system from Robot MarketPlace Twin Stick — Futaba T9CAP radio and receiver

with PCM and PPM, nine channels.

ly limited and has a lot of

interference It should never be

used on a bot that has a weapon

FM radios are better than AM —your radio will have less interference

The middle range FM radios are PPM

which is an analog system The

better FM radios have what is calledPCM which is a digital system TheFM-PPM radio can also have PCM

The difference between the two is inhow the signal is encoded PCM signals are encoded digitally andgive a higher degree of immunity

from noise than PPM The best way

to find out the type of radio thatwould work best for you is to talk toother combat robot builders With agood radio choice in the beginning,you will definitely be able to use it inmore than one robot SV

This was the first charger I

bought after I got into the

sport, at the recommendation of a

mentor who came up through the

R/C car racing side of things Priced

new at $64.99, I picked it up on

eBay for about $35 I’ve found it to

be a very dependable, basic charger

that’s handled my NiMH and NiCad

charging needs very well

The model DTX4110 Mini Pulsecharger comes with a handy,

detachable, 12V power supply that

provides seven amps This has come

in handy, both as a quickie benchsupply, and the provided alligatorclips from the charger itself lets mecharge packs from a small sealedlead acid battery in the van on theway to events

The Mini Pulse can charge up tothree amps for up to eight NiMHcells NiCads can be charged at up

to 4.5 amps There is a two-amp constant discharge function which,

to be honest, I’ve never used since

my packs get discharged in driving practice

PRODUCT REVIEW — DuraTrax IntelliPeak

AC/DC Mini Pulse Charger

● by Kevin Berry

ComBots Cup 2007 — This event will take place on1/14-15/2007 in Oakland, CA There is a $10,000Heavyweight prize and a $3,000 Middleweight prize

Visit www.robo

games.net formore details

Bay Area RobotFights — This eventwill take place on1/27/2007 in St

Petersburg, FL It is thefourth event in thisannual series — a conventional insect battle run by somevery unconventional people Fun for the whole family.This event data is tentative at time of publication SV

EVENTS

UPCOMING — December

and January

The promise of collaborative

robot-ics is a new breed of machines that

are freed from the role of obedient

automaton or tele-robot in need of

constant supervision Instead, they can

engage in a dialogue with humans, ask

and answer questions, and resolve

differences in order to achieve shared

goals As a result, humans are free to

focus on more important tasks, only

occasionally interrupted by robots in

need of assistance to compensate for

their limited autonomy

In practice, creating robots capable

of collaborating with humans is no

mean feat For example, not only must

a collaborative robot have the ability to

act autonomously, but it must be able

to modulate the level of autonomy to

suit the situation Furthermore, there is

the human element to consider It will

take time for people to accept the idea

of working with — or even for — a robot

This article introduces collaborative

robotics and discusses several key

implementation challenges that

read-ers should consider as they explore this

frontier of robotics

Collaboration

A working definition of

collabora-tion incorporates the concept of two or

more entities working together to

achieve a shared goal more efficiently

and/or effectively than would be

possi-ble by working independently A

com-mon feature of collaboration includes

either explicit or implied rank or status

and a corresponding deference to the

collaborator in authority Collaboration

is often fostered if collaborators canpredict or anticipate the needs of othercollaborators, either by task assign-ment or through experiential learning

Collaboration isn’t necessarily agood thing or even desirable in everysituation — there are instances where asingle, autonomous individual of action

is preferable to a committee of sive decision makers However, thereare numerous examples of situations inwhich efficiency and effectivenessgains can be realized through collabo-ration, such as hunting in a pack, teamsports, warfare, and firefighting

indeci-The most publicized example ofrobot-robot collaboration is the annualRoboCup competition, which is aninternational proving ground for collab-orative robotics and related AI topics

[1] The goal of the organization is todevelop a team of fully autonomoushumanoid robots that can win againstthe human world soccer championshipteam by 2050

Developments in the more challenging frontier of human-robotcollaboration are represented byNASA’s Robonaut [2] and, to a lesserextent, DARPA’s BigDog [3] Robonaut

is NASA’s attempt to create a roboticastronaut capable of working side-by-side with astronauts in the InternationalSpace Station DARPA’s BigDog — arobotic replacement for pack animalsthat is capable of carrying a 100-poundpayload over rugged terrain — sharescharacteristics of both traditional ‘slave’robots and collaborative robots

Collaborative robots typically rely

on multiple, powerful computer

The normal course of human social and cognitive development begins with total dependence at birth and progresses to the semi-autonomy of adolescence Most successful adults advance to a stage of collaborative interdependence that acknowledges the benefits of working with others toward a common goal This social evolution is reflected in robotics,

where there is increased interest in developing robots that can not only

work with each other, but work collaboratively with humans.

processors and algorithms Even so,

biological organisms suggest that

heavy iron isn’t a prerequisite for robot

collaboration Ants, bees, and many

other insects rely on distributed or

swarm intelligence to accomplish

collaborative feats that researchers

have only begun to approximate [4]

Robot-Robot

Collaboration

The simplest form of collaborative

robotics is collaboration among two

robots However, achieving a true

collab-orative relationship is far from simple To

appreciate the challenge of designing a

pair of robots capable of collaborating

on a straightforward task, consider the

interaction depicted in Figure 1 Each

robot is equipped with a typical suite of

autonomous features, including the

ability to avoid obstacles, locate objects

by color or shape, manage energy

stores, and navigate Furthermore, each

robot is programmed to roam and

acquire balls that come into sensor

range That is, they have an identical —

but not shared — goal

Returning to Figure 1, assume the

robot with the green dome is first to

detect the ball It establishes a direct

path to the target and, as it

approach-es, it monitors the robot-wall distance

yellow dome detects the ball at aboutthe same time, but it has a more ardu-ous task to get into position It movesthrough the doorway and heads towardthe target However, neither rapidly-moving robot is adequately equippedwith sensors or programming to sensethe other robot in time The result is acollision that may damage both robots

How could the collision have beenavoided? More importantly, how can therobots work collaboratively to acquireballs as effectively as possible? One solu-tion is communications Given adequaterobot-robot communications, the firstrobot to locate a ball can signal the otherrobot that it is preparing to capture theball The second robot would then befree to search for other balls and avoidinterfering with the first robot This veryloose collaboration consists primarily ofrobots staying out of each other’s way

Another solution involves ing the sensors so that robots candetect each other from a significantdistance and establishing one robot

enhanc-as the ‘alpha’ robot Given potentialcompetition over a target, the secondrobot defers to the alpha robot

Collaboration can also be enabled

by assigning different roles to eachrobot For example, one robot isassigned the role of scout, and theother the role of retriever Role assign-

heavy, power-hungry, and costly sors, while the retriever can be outfit-ted with fewer and less costly sensors.Role assignment, while seeminglystraightforward, has a number ofdependencies The first is task decomposition which, in turn, definesnecessary robot competencies A robottasked with scouting for possible tar-gets may spend most of its time roam-ing, following walls, entering doorways,and perhaps periodically notifying theother robot of its position The mostimportant scout robot competenciesare related to spotting targets from asignificant distance, mapping the area,and communicating the target coordi-nates — and perhaps even the most efficient route — to the retriever

sen-In contrast, the retriever mightspend most of its time in an energy-saving mode, awaiting a signal fromthe scout before activating its drive sys-tem and navigation sensors Its primarycompetencies involve receiving andinterpreting communications from thescout, following the suggested path tothe target from its current location,and efficiently acquiring the target.The competencies of both robotsmust be supported by the appropriatesensors, effectors, hardware platform,and operating system Borrowing atechnique from game development, it’shelpful to create a robot profile foreach robot, as in Figure 2 The profileshould contain graphical and tabulardata on sensor range and function andbasic physical capabilities of each robot

In this simplified profile, the robot is amodest wheeled robot, relatively low tothe ground, and equipped with anomni-camera, IR laser rangefinder, aswell as discrete US and IR sensors

Tight Collaboration

In the previous example of a loosecollaboration, robots needn’t evercome in close proximity to each other,except by accident In contrast, withtight collaboration, robots work interac-tively to achieve a shared goal Considerthe scenario depicted in Figure 3, inwhich a pair of robots collaborate inhunting down a human player in agame of laser tag The shared goal is to

FIGURE 2.

Simplified

robot profile

(not to scale).

As demonstrated by say, the pursuit

of a deer by a pair of wolves,

collabora-tive pursuit is a complex, carefully

choreographed interaction in which the

smallest mistake on the part of the

pred-ator can shift the advantage to the prey

Whereas a single predator may approach

its prey from behind and at a slight angle

to encourage the prey to follow an

inefficient arc in an attempt to escape,

two predators have many more options

One predator can wait in ambush, while

the other predator maneuvers the prey

into position Alternatively, one predator

can distract the prey while the other

predator takes up a superior position

The two predators could opt to simply

surprise and confuse the prey by

appear-ing at the same instant

In each case, the coordinated

behaviors require planning,

communi-cations, and the ability to predict not

only how the prey will respond to a

threat, but how the collaborating robot

will respond to changes in prey

behav-ior Prediction is an important

capabili-ty, and one not easily satisfied without

an internal model of how collaborating

robots will react to potentially novel

situations Predicting human prey

behavior is particularly challenging

In a tight collaboration scenario,

the robot profiles are crucial for

under-standing the interplay of the robot

sensors with the environment, the

target, and the sensor signals from

collaborating robots For example, the

two robots approaching the entrance

of a room holding the hostile in Figure

3 are painting each other with

ultra-sound and IR signals Even though the

ranging distance might be limited to a

meter or two, the transmitted IR and US

signals travel hundreds of meters and

reflect off of walls and other structures

The result is an increase in the noise

floor, which appears as random

varia-tion in distance measures Furthermore,

if the ultrasound sensors are receptive

to reflected signals, a sensor on one

robot can be triggered by the direct

or reflected signal from another

transducer As the number of robots

increases, so does the noise and risk of

sensor saturation and false triggering

Human-Robot Collaboration

Human-robot collaboration — for

years limited to the realm of science fiction — is the real frontier of robotics Inaddition to the capabilities required forrobot-robot collaboration, the human

FIGURE 3 Snapshot

of collaborative team sensor coverage profile To reach the target, robots must pass into the room through a doorway.

FIGURE 4 Human-robot collaboration

in a game of laser tag involving

coordinated entry into a room with

a hostile (blue shirt).

element demands new ways of thinking

about robots and of human relationships

Figure 4 depicts a scenario in

which a team of three robots and one

human (white shirt) are playing a game

of laser tag against another human

(blue shirt) At first glance, neutralizing

the outnumbered hostile is simply a

matter of entering the room,

identify-ing the hostile, and firidentify-ing

However, on closer examination,

not only is entering the room without

creating a pileup in the doorway a

high-ly complex maneuver, but simphigh-ly

assem-bling in an orderly fashion outside the

door is beyond the capabilities of most

autonomous robots As in two-robotcollaboration, role assignment, taskdecomposition, and modeling collabora-tor and prey responses are critical

Simply adding a third robot to theteam increases complexity exponential-

ly For example, consider what’sinvolved in expanding intra-team communications from two to threerobots With two collaborating robots,there is no need for a robot to identifyitself or the intended recipient of amessage With three robots, there ismessage ambiguity without a means ofidentifying the source of each messagebecause messages can be generated by

one of two other robots Furthermore,communications can be broadcast tothe other two robots or, given theproper communications equipment, arobot can be singled out to receive aprivate message Robots must also con-tend with the noise and interferencefrom three robots in close proximitygenerating audio, RF, US, and IR noise.Let’s explore the challenge of walking in formation toward the roomentrance An autonomous robot simplycontends with following the wall, identi-fying the entrance, and turning into thedoorway as soon as the edge of theentrance is recognized Moving in for-mation requires a new set of behaviorsand competencies related to leading,following, robot-robot and robot-humanspacing, and velocity modulation

As shown in Figure 5, one scenario

is for the team leader — in this case, ahuman — to initiate the approach Thefirst robot (green dome) reacts to thismovement by accelerating toward theleader, and the second robot (reddome) similarly reacts to the move-ment of the first robot However,depending on the acceleration charac-teristics of the human and robots,response time of sensors, and targetspacing, the formation can move inunison, like an oscillating slinky, or thefirst robot could ram into the ankle

of the lead human and the secondrobot follow up by crashing into thefirst robot

The velocity plot shown in Figure 5illustrates a responsive first robot thattracks human velocity with a lag ofabout one second It both over-andunder-shoots the human’s velocity Thesecond robot responds poorly to thefirst robot’s movements, and has difficulty maintaining a constant robot-robot distance The solutionmight involve equipping the secondrobot with a more capable quadencoder and PID configuration and perhaps replacing brushed motors with

a more predictable brushless design.Designing robots capable of supporting human-robot collaboration

is especially challenging when the team

is intended to interact with humans Ahuman target in the laser tag gameintroduces the concept of friend versus

5RERW

FIGURE 5 Velocity plot associated with

a collaborative human-robot team walking in formation toward a target.

FIGURE 6 2D IR laser rangefinder plot

of a person standing

in an open doorway, similar to the scenario

in Figure 4 Created with a Hokuyo URG- 04LX mounted on a robot about 15 cm above ground level.

may require reinterpreting,

augment-ing, or completely replacing sensors

For example, the IR laser used to

identify the human target in the

robot-robot collaboration example (Figure 3)

is inadequate for differentiating

between the human team member

and the human hostile

As shown in Figure 6, a human

friend or foe facing a relatively short

robot equipped with an IR laser

rangefinder appears as two shadows —

one for each leg It’s even more

chal-lenging to identify a human standing

perpendicular to the robot, since the

legs produce only a single shadow In

either case, differentiating friend from

foe based on leg shadows alone isn’t

possible One solution is to configure

the omnidirectional camera to

recog-nize colors, assuming the hostile

and team member consistently wear

blue and white shirts Fusing the IR

laser rangefinder and color data could

provide a more accurate indication of

friend and foe position

Another challenging task is

track-ing the lead human in the initial lineup

prior to entering the room with the

hostile The laser rangefinder used to

create Figure 6 provides a 240 degree

field-of-view, updated at 10 Hz While

this refresh rate may be sufficient for

mapping purposes, it is inadequate for

tracking the rapidly moving feet of the

lead human, especially if the human

stops or turns unexpectedly

There is also the issue of team

communications Candidates for

human-robot communications range

from a simple, multi-button RF

signal-ing device to more computationally

intensive gesture or voice recognition

Performance constraints are also

critical in human-robot collaboration

Although the relationship will

eventual-ly evolve, current performance

standards are imposed by humans A

human can travel and respond faster to

visual cues than any commercial robot

In an operation such as depicted in

Figure 4, the robots limit the pace of

the operation

From Here

Collaborative robotics is beginning

to appear in military and even some

commercial offerings For example,

robotic surgical assistants now

routine-ly assist surgeons performing prostatesurgery in hospitals throughout the US

The caveat is that the relationship is thetypical master-slave relationship of old

— which makes sense, given that surgeons aren’t likely to invest in technology intended to replace them —

a fear highlighted by Ellison’s classic

short story Wanted in Surgery [5].

This fear of robots empoweredwith collaborative leadership qualities isbeing addressed head-on by research inthe area of sociable robotics Althoughlimited to research experiments, sociable robots have demonstrated thevalue of robots that can understandand relate to humans in a personalway

One of the best known sociablerobots — MIT’s Kismet [6] — illustratesthe limitation of current technology

Kismet is more animatronic than robotic, in part because it relies on anetwork of 15 PCs to enable it to exhib-

it a modicum of social intelligence Intime, multi-core parallel processing andnew algorithms will bring socially intelligent, collaborative behaviors to

every robot For now, collaborativerobotics is a practical goal that is wellwithin the reach of every enthusiastwilling to take up the challenge SV

[1] Nardi, D., M Riedmiller, et al., eds.

RobCup 2004: Robot Soccer World Cup VIII 2005, Springer: New York.

[2] Ambrose, R Robonaut 2006 [cited

2006 June 13]; Available from: vesu vius.jsc.nasa.gov/er_er/html/robonaut /robonaut.html.

[3] Hambling, D Robotic ‘pack mule’

displays stunning reflexes 2006 [cited

2006 May 15]; Available from: www.new scientist.com/article.ns?id=dn8802 [4] Tarasewich, P and P McMullen,

Swarm intelligence Communications of the ACM, 2002 45(8): p 62-7.

[5] Ellison, H., Wanted in Surgery, in If.

1957, Quinn Publishing Company: Buffalo, NY Audio version available on

Audible.com.

[6] Breazeal, C., Designing Sociable

Robots 2002: MIT Press.

REFERENCES

ELEXOL Ether I/O 24

• Supports ARP, BOOTP, DHCP, ICMP and UDP/IP protocols

• Standard 10BaseT Ethernet Interface; RJ45 connector

• 24 independently programmable signal lines with confi gurable CMOS, TTL or Schmitt Trigger thresholds and programmable pull-ups per line

• Easy connection by three 10-way box headers

• On-board 50MIPS fl ash micro-controller may be reprogrammed to suit specifi c applications

• Integrated switch-mode voltage regulator allows power supplies from 8-32VDC

• User 5V 500mA output to power external interface boards or sensors

• Compact module – 72 x 72 x 24mm

CONNECTOR/LED BOARD

Provides screw terminal connections plus optional either pull up or pull down resistors and LED port status indication.

Great range of I/O-24 peripherals, too

OPTO INPUT BOARD

Provides 8 Opto isolated inputs for each of the I/O-24 pins with LED indications.

RELAY OUPUT BOARD

Provides 8 isolated relay contact outputs suitable for a variety of loads.

SWITCH/PUSH BUTTON BOARD

Provides 8 switch/push buttons with LED indicators showing the status of the I/O pin.

Visit our website www.orteches.com

Ortech Education Systems

1119 - 9th Ave S • Moorhead, MN 56560

218.287.1379

UDP/IP controlled digital input/output module featuring three 8-bit ports with 5V level signal lines Each of the 24 lines can be independently programmed as either an input

or an output Connects to any TCP/IP protocol network.

www.orteches.com www.orteches.com www.orteches.com www.orteches.com www.orteches.com www.orteches.com

www.orteches.com www.orteches.com www.orteches.com www.orteches.com www.orteches.com www.orteches.com

Perform proportional speed, direction, and steering with

only two Radio/Control channels for vehicles using two

separate brush-type electric motors mounted right and left

with our mixing RDFR dual speed control Used in many

successful competitive robots Single joystick operation: up

goes straight ahead, down is reverse Pure right or left twirls

vehicle as motors turn opposite directions In between stick

positions completely proportional Plugs in like a servo to

your Futaba, JR, Hitec, or similar radio Compatible with gyro

steering stabilization Various volt and amp sizes available

The RDFR47E 55V 75A per motor unit pictured above

www.vantec.com

STEER WINNING ROBOTS

WITHOUT SERVOS!

Order at (888) 929-5055

Extreme Robot Speed Control!

OSMC - Monster Power H-bridge

6 6 6 6 6

14V - 50V and 3.15“ x 4.5” x 1.5”

Control with Stamp or other Micro

3 wire interface R/C interface available

160Aover400Apeak!

Dual (6A pk) H-bridges fwd-only channel 5V - 18V

1.6“ x 1.6” x 0.5”

Four R/C inputs Mixing, Flipped Bot Input

2.5A 12A

Plus

$79.99 Scorpion Mini

6 6 6 6 6 6

2.5A (6A pk) H-bridge

5V - 18V 1.25“ x 0.5” x 0.25”

Control like a servo Optional screw term.

Only 5.5g

$119.99 Scorpion XL

6 Only 28g

6 6 6 6 6

Dual H-bridge 5V - 24V 2.7“ x 1.6” x 0.5”

Three R/C inputs - serial option Mixing, Flipped Bot Input

13A 45A Peak!

6 6 6 6 6 6 6 6 6 6

14V - 50V Dual H-bridges 150 Adjustable current limiting Adjustable speed slew rate Temperature limiting Three R/C inputs - serial option Many mixing options Flipped Bot Input Rugged extruded Aluminum case 4.25" x 3.23" x 1.1” - Only 365g

80A A+ Peak!

$29.99

$399

Also from Robot Power

Kits, parts, schematics Planetary gearmotors

All Robot Power electronicproducts are proudly

Introducing Dalf

Advanced dual motordrive with closed-loopcontrol functions

Embedded Electronics, LLC along with our exclusive reseller Robot Power are proud to introduce a feature rich, customizable Dual Motor Controller: Designed to work out of the box or to host your application specific code; Dalf makes it simple to create a complete turn-key “brain” for your application with full-closed-loop motion control Just take a look at these features!

6 6 6 6 6 6 6 6 6 6 6

Closed-loop control of two motors Full PID position loop Trapezoidal path generator Adjustable slew rate for smooth transitions Non-volatile storage of PID parameters Step-Response PID motor tuning support Quadrature encoder support for each motor

Drives all sign-magnitude brushed DC motor drives such as the OSMC

Terminal mode for interactive tuning and debugging

Windows GUI under development

Two R/C command modes (3 input channels)

Two open-loop pot control modes Interactive terminal control of motors Adjustable slew rate

6

Open-Loop Features

I/O Connections

6 6 6 6 6 6 6

Two RS-232 serial ports

36 GPIO I2C master and slave ports (2 ports) Two motor drive outputs Two quadrature encoder inputs Two Hall-effect current sensors inputs Six 10-bit A/D

Two channels of cooling fan control

6 Standard ICD connector

Application Support

6 6 6 6 6 6 6 6 6

60k+ FLASH available Serial bootloader, no programmer needed Serial command/monitor in both terminal and high-speed binary API mode I2C slave command interface Firmware implented in C andASM

C source for main loop and utility routines provided free

Linkable device driver function library provided for building custom applications Extensive documentation with Owner’s Manual and Getting Started Manual provided on CD

Custom code development services available (contact EE)

6 PIC18F6722 CPU running at 40MHz

For more information visit www.embeddedelectronics.net

Dalf

DARwIn (Dynamic Anthropomorphic Robot

with Intelligence) is a humanoid robot

capable of bipedal walking and performing

human-like motions Developed at the Robotics &

Mechanisms Laboratory (RoMeLa) at Virginia Tech,

DARwIn is a research platform for studying robot locomotion

and also the base platform for Virginia Tech’s first entry to the

2006 Robocup competition (Figure 1) First started as a design

project for seniors in the department of mechanical engineering

at Virginia Tech, the general goal in designing DARwIn was to

create a bipedal robot with human proportions; suitable to be a researchplatform capable of dynamic walking The 600 mm tall, 4 Kg robot has

21 degrees-of-freedom (DOF) with each joint actuated by coreless DCmotors via distributed control with controllable compliance Using a computer vision system on the head, rate gyros in the torso, and multipleforce sensors on the foot, DARwIn will be able to implement human-likedynamic gaits while navigating obstacles and traverse uneven terrain whileimplementing complex behaviors such as playing soccer

This three-part series will describe the design and fabrication processfor the first version of DARwIn’s hardware, highlight the salient features,

PART 1:

Concept and General Overview

by: Karl Muecke, Patrick

Cox, and Dennis Hong

RoMeLa (Robotics & Mechanisms

Lab) at Virginia Tech;

www.me.vt.edu/romela

< FIGURE 1.

Robocup 2006 participants (DARwIn not pictured) Photo courtesy of Messe Breme www robocup 2006.org.